A peer-reviewed version of this preprint was published in PeerJ on 15 June 2018. View the peer-reviewed version (peerj.com/articles/5067), which is the preferred citable publication unless you specifically need to cite this preprint. Hernandez AM, Ryan JF. 2018. Horizontally transferred genes in the ctenophore Mnemiopsis leidyi. PeerJ 6:e5067 https://doi.org/10.7717/peerj.5067 Horizontally transferred genes in the ctenophore Mnemiopsis leidyi Alexandra M Hernandez 1, 2 , Joseph F Ryan Corresp. 1, 2 1 Whitney Laboratory for Marine Bioscience, St. Augustine, Florida, United States 2 Department of Biology, University of Florida, Gainesville, FL, United States Corresponding Author: Joseph F Ryan Email address: [email protected] Horizontal gene transfer has had major impacts on the biology of a wide range of organisms from antibiotic resistance in bacteria to adaptations to herbivory in arthropods. A growing body of literature shows that horizontal gene transfer (HGT) between non- animals and animals is more commonplace than previously thought. In this study, we present a thorough investigation of HGT in the ctenophore Mnemiopsis leidyi. We applied tests of phylogenetic incongruence to identify nine genes that were likely transferred horizontally early in ctenophore evolution from bacteria and non-metazoan eukaryotes. All but one of these HGTs (an uncharacterized protein) are homologous to characterized enzymes, supporting previous observations that genes encoding enzymes are more likely to be retained after HGT events. We found that the majority of these nine horizontally transferred genes were expressed during development, suggesting that they are active and play a role in the biology of M. leidyi. This is the first report of HGT in ctenophores, and contributes to an ever-growing literature on the prevalence of genetic information flowing between non-animals and animals. PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3462v3 | CC BY 4.0 Open Access | rec: 6 Jun 2018, publ: 6 Jun 2018 1 Horizontally transferred genes in the ctenophore Mnemiopsis leidyi 2 3 Alexandra M. Hernandez1,2 and Joseph F. Ryan1,2 4 5 1 Whitney Laboratory for Marine Bioscience, University of Florida, St. Augustine, FL, USA 6 2 Department of Biology, University of Florida, Gainesville, FL, USA 7 8 Corresponding Author: 9 Joseph Ryan1 10 [email protected] 11 Abstract 12 13 Horizontal gene transfer has had major impacts on the biology of a wide range of organisms 14 from antibiotic resistance in bacteria to adaptations to herbivory in arthropods. A growing body 15 of literature shows that horizontal gene transfer (HGT) between non-animals and animals is more 16 commonplace than previously thought. In this study, we present a thorough investigation of HGT 17 in the ctenophore Mnemiopsis leidyi. We applied tests of phylogenetic incongruence to identify 18 nine genes that were likely transferred horizontally early in ctenophore evolution from bacteria 19 and non-metazoan eukaryotes. All but one of these HGTs (an uncharacterized protein) are 20 homologous to characterized enzymes, supporting previous observations that genes encoding 21 enzymes are more likely to be retained after HGT events. We found that the majority of these 22 nine horizontally transferred genes were expressed during development, suggesting that they are 23 active and play a role in the biology of M. leidyi. This is the first report of HGT in ctenophores, 24 and contributes to an ever-growing literature on the prevalence of genetic information flowing 25 between non-animals and animals. 26 Introduction 27 28 Evolution is commonly thought to occur by descent with modification from a single 29 lineage. However, evidence has shown that genomes from bacteria, archaea, and eukaryotes are 30 typically chimeric, resulting from horizontal (or lateral) gene transfers (Garcia-Vallvé et al. 31 2000; Katz 2002). As such, horizontal gene transfer (HGT) has likely impacted evolution more 32 than originally thought by creating opportunities for rapid genetic diversification and 33 contributing to speciation events. Moreover, HGT is a potential catalyst for organisms to acquire 34 novel traits (Soucy et al. 2015) and creates opportunities for HGT receivers to exploit new 35 ecological niches (Boto 2010). For example, HGTs have played an important role in herbivory in 36 arthropods (Wybouw et al. 2016), venom recruitment in parasitoid wasps (Martinson et al. 37 2016), cellulose production in urochordates (Dehal et al. 2002) and plant parasitism in 38 nematodes (Haegeman et al. 2011). 39 40 Although HGT is generally accepted as an important evolutionary mechanism in 41 prokaryotes (Boto 2014), it remains controversial whether it occurs in animals, despite many 42 convincing studies (Madhusoodanan 2015). Much of the skepticism has been fueled by high- 43 profile reports of HGT (e.g., Lander et al. 2001; Boothby et al. 2015) that were later shown to be 44 largely incorrect due to contamination or taxon sampling (Stanhope et al. 2001; Koutsovoulos et 45 al. 2016). In addition, HGT in animals is hypothesized to be rare due to the origin of a 46 sequestered germ line, which provides fewer opportunities for germ cells to be exposed to 47 foreign DNA (Doolittle 1999; Andersson et al. 2001; Jensen et al. 2016). However, the presence 48 and absence of germline sequestration is not well described across the animal tree of life, and 49 there are inconsistencies between studies regarding which animal groups have sequestered 50 germlines (Buss, 1983; Radzvilavicius et al. 2016; Jensen et al. 2016). 51 52 The major challenges for HGT detection efforts have been taxon sampling and 53 contamination. Many early reports of HGT in animals were overturned due to limited 54 representation of taxa in public genomic databases (e.g., Salzberg et al. 2001). For example, a 55 gene present in bacteria and humans, but absent from nematodes and drosophilids (the most 56 highly represented taxa at the time) may have been considered the result of HGT, until 57 discovering that the gene is present in many other animal genomes that were not available at the 58 time of the initial claim. In these cases, the limited representation of taxa made it difficult to 59 distinguish HGTs from differential gene loss (Andersson et al. 2006; Keeling & Palmer 2008). 60 More recently, contamination has led to both overestimation and likely underestimation of HGT 61 events. In several recent cases, contamination in newly generated datasets has been interpreted as 62 HGT but later shown to be cross-contaminants present in genome sequences (Bhattacharya et al. 63 2013; Delmont & Eren 2016; Koutsovoulos et al. 2016). On the other hand, the presence of 64 contaminants in public databases (e.g., a bacteria sequence labeled as an animal sequence) makes 65 it difficult to identify bona fide HGTs, as “animal” sequences will appear among the top BLAST 66 hits for a particular HGT, leading to false negatives (Kryukov & Imanishi 2016). As such, 67 contamination remains a major hurdle to contemporary studies of HGT. 68 69 Pairwise BLAST-based similarity scores (e.g., alien index (Gladyshev et al. 2008) and 70 the HGT index (Boschetti et al. 2012)) are the most common criteria used to detect HGT in 71 animals. However, these measures largely ignore phylogenetic information associated with 72 sequence data. While a positive BLAST-based result may be due to HGT, it may also result from 73 gene loss, selective evolutionary rates, convergent evolution, sequence contamination, and 74 species misassignment (Hall et al. 2005). Previous HGT studies have demonstrated that HGT 75 predictions need to be carefully considered and a combination of methods are required to rule out 76 false positives (Schönknecht et al. 2013). Hypothesis tests incorporating phylogenetic 77 incongruence are one such method that has been used to test HGT. While some studies in 78 animals have incorporated these techniques (e.g., Eliáš et al. 2016), they are more commonly 79 deployed in studies involving non-animals (e.g., Bapteste et al. 2003; Richards et al. 2006). 80 81 HGT has yet to be thoroughly explored in Ctenophora. Ctenophores (comb jellies) are 82 marine invertebrates that are morphologically characterized by eight rows of cilia used for 83 movement. They typically live in the water column, but the group includes benthic species as 84 well (Song & Hwang 2010; Alamaru et al. 2015; Glynn et al. 2017). Phylogenomic evidence 85 from studies including ctenophores has suggested that ctenophores are the sister group to all 86 other animals (Dunn et al. 2008; Hejnol et al. 2009; Ryan et al. 2013; Moroz et al. 2014; 87 Borowiec et al. 2015; Chang et al. 2015; Torruella et al. 2015; Whelan et al. 2015; Arcila et al. 88 2017; Shen et al. 2017; Whelan et al. 2017), but the position remains controversial with some 89 evidence supporting sponges as the sister group to the rest of animals (Philippe et al. 2009; Pick 90 et al. 2010; Pisani et al. 2015; Telford et al. 2015; Simion et al. 2017; Feuda et al. 2017). Thus, 91 investigating HGT in ctenophores is essential to understanding its implications on early animal 92 evolution. 93 94 Here, we apply a rigorous framework to identify and confirm HGTs in the ctenophore 95 Mnemiopsis leidyi. Our process includes identification of HGT candidates by alien index and 96 confirmation by phylogenetic hypothesis testing, providing statistical support in an evolutionary 97 framework. Furthermore, we analyze gene expression profiles during development to obtain 98 clues as to the function of these HGTs in M. leidyi. 99 100 Material and Methods 101 102 All command lines, parameters, and version numbers of programs are in the supplementary text. 103 104 Identification of HGT candidates by alien_index 105 106 As part of this project, we developed the program alien_index and complimentary 107 metazoan/non-metazoan sequence databases to automate the generation of alien index 108 (Gladyshev et al.